import pandas as pd
import numpy as np
import re
import statsmodels.api as sm
import statsmodels.formula.api as smf
from statsmodels.stats.outliers_influence import variance_inflation_factor
from statsmodels.stats.diagnostic import het_breuschpagan, normal_ad
from scipy import stats

# reproducible random for assigning random floors
RANDOM_SEED = 42
rng = np.random.default_rng(RANDOM_SEED)

df = pd.read_csv('clean_variable_all.csv')

df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 4665 entries, 0 to 4664
Data columns (total 58 columns):
 #   Column                          Non-Null Count  Dtype  
---  ------                          --------------  -----  
 0   Area (in Sq. ft)                4665 non-null   float64
 1   Amount                          4665 non-null   float64
 2   Market Value                    4665 non-null   float64
 3   City                            4665 non-null   object 
 4   Project Name                    4665 non-null   object 
 5   Location                        4665 non-null   object 
 6   Latitude                        4665 non-null   float64
 7   Longitude                       4665 non-null   float64
 8   Sale Type                       4665 non-null   object 
 9   Registration Date               4665 non-null   object 
 10  Quaters                         4665 non-null   object 
 11  YR                              4665 non-null   int64  
 12  Amount in CR                    4665 non-null   float64
 13  Price Range                     4665 non-null   object 
 14  Project Name main               2306 non-null   object 
 15  Rate (Rs/sft)                   4665 non-null   object 
 16  Unit                            4301 non-null   object 
 17  Floor                           4506 non-null   object 
 18  Tower                           962 non-null    object 
 19  Wing                            1438 non-null   object 
 20  Party 1                         4665 non-null   object 
 21  Party 2                         4665 non-null   object 
 22  Count                           4665 non-null   object 
 23  bus_stop_count_1km              4665 non-null   int64  
 24  financial_facilities_1km        4665 non-null   int64  
 25  medical_facilities_1km          4665 non-null   int64  
 26  hospital_facilities_1km         4665 non-null   int64  
 27  distance_to_metro               4665 non-null   float64
 28  distance_to_railway             4665 non-null   float64
 29  school_count_1km                4665 non-null   int64  
 30  kindergarten_count_1km          4665 non-null   int64  
 31  distance_to_supermarket         4665 non-null   float64
 32  distance_to_techhub             4665 non-null   float64
 33  distance_to_park                4665 non-null   float64
 34  distance_to_waterbody           4665 non-null   float64
 35  park_name                       750 non-null    object 
 36  park_area_m2                    4665 non-null   float64
 37  park_lat                        4665 non-null   float64
 38  park_lon                        4665 non-null   float64
 39  waterbody_name                  3080 non-null   object 
 40  waterbody_waterbody_type_clean  4665 non-null   object 
 41  waterbody_area_m2               4665 non-null   float64
 42  waterbody_lat                   4665 non-null   float64
 43  waterbody_lon                   4665 non-null   float64
 44  Floor_clean                     4665 non-null   int64  
 45  SaleType_encoded                4665 non-null   int64  
 46  park_leisure_garden             4665 non-null   bool   
 47  park_leisure_park               4665 non-null   bool   
 48  park_leisure_pitch              4665 non-null   bool   
 49  park_leisure_playground         4665 non-null   bool   
 50  park_leisure_recreation_common  4665 non-null   bool   
 51  waterbody_type_lake             4665 non-null   bool   
 52  waterbody_type_pond             4665 non-null   bool   
 53  waterbody_type_reservoir        4665 non-null   bool   
 54  waterbody_type_river            4665 non-null   bool   
 55  log_amount                      4665 non-null   float64
 56  area_raw                        4665 non-null   float64
 57  log_area                        4665 non-null   float64
dtypes: bool(9), float64(21), int64(9), object(19)
memory usage: 1.8+ MB

# Log of Amount (dependent variable)
df['log_amount'] = np.log(df['Amount'])
# Option A: keep Area as raw
df['area_raw'] = df['Area (in Sq. ft)']

# Option B: also create log-area (to try log-log model later)
df['log_area'] = np.log(df['Area (in Sq. ft)'])

# Choose predictors
X = df[['log_area', 'Floor_clean', 'SaleType_encoded',
        'bus_stop_count_1km', 'financial_facilities_1km',
        'medical_facilities_1km',
        'distance_to_metro', 'distance_to_railway', 'school_count_1km',
        'kindergarten_count_1km', 'distance_to_supermarket',
        'distance_to_techhub', 'distance_to_park', 'distance_to_waterbody',
        'park_area_m2', 'waterbody_area_m2',
        'park_leisure_garden', 'park_leisure_park',
       'park_leisure_pitch', 'park_leisure_playground',
       'park_leisure_recreation_common',
       'waterbody_type_lake',
       'waterbody_type_pond', 'waterbody_type_reservoir',
       'waterbody_type_river']]

# Find all boolean columns
bool_cols = X.select_dtypes(include=['bool']).columns
print(bool_cols)
# Convert all boolean columns to integers
X[bool_cols] = X[bool_cols].astype(int)

Index(['park_leisure_garden', 'park_leisure_park', 'park_leisure_pitch',
       'park_leisure_playground', 'park_leisure_recreation_common',
       'waterbody_type_lake', 'waterbody_type_pond',
       'waterbody_type_reservoir', 'waterbody_type_river'],
      dtype='object')

C:\Users\Sayan\AppData\Local\Temp\ipykernel_20080\2290739045.py:5: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  X[bool_cols] = X[bool_cols].astype(int)

X.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 4665 entries, 0 to 4664
Data columns (total 25 columns):
 #   Column                          Non-Null Count  Dtype  
---  ------                          --------------  -----  
 0   log_area                        4665 non-null   float64
 1   Floor_clean                     4665 non-null   int64  
 2   SaleType_encoded                4665 non-null   int64  
 3   bus_stop_count_1km              4665 non-null   int64  
 4   financial_facilities_1km        4665 non-null   int64  
 5   medical_facilities_1km          4665 non-null   int64  
 6   distance_to_metro               4665 non-null   float64
 7   distance_to_railway             4665 non-null   float64
 8   school_count_1km                4665 non-null   int64  
 9   kindergarten_count_1km          4665 non-null   int64  
 10  distance_to_supermarket         4665 non-null   float64
 11  distance_to_techhub             4665 non-null   float64
 12  distance_to_park                4665 non-null   float64
 13  distance_to_waterbody           4665 non-null   float64
 14  park_area_m2                    4665 non-null   float64
 15  waterbody_area_m2               4665 non-null   float64
 16  park_leisure_garden             4665 non-null   int32  
 17  park_leisure_park               4665 non-null   int32  
 18  park_leisure_pitch              4665 non-null   int32  
 19  park_leisure_playground         4665 non-null   int32  
 20  park_leisure_recreation_common  4665 non-null   int32  
 21  waterbody_type_lake             4665 non-null   int32  
 22  waterbody_type_pond             4665 non-null   int32  
 23  waterbody_type_reservoir        4665 non-null   int32  
 24  waterbody_type_river            4665 non-null   int32  
dtypes: float64(9), int32(9), int64(7)
memory usage: 747.3 KB

X_1 = X[['log_area', 'Floor_clean', 'SaleType_encoded',
        'bus_stop_count_1km', 'financial_facilities_1km',
        'medical_facilities_1km',
        'distance_to_metro', 'distance_to_railway', 'school_count_1km',
        'kindergarten_count_1km', 'distance_to_supermarket',
        'distance_to_techhub','distance_to_park', 'distance_to_waterbody',
        'park_area_m2', 'waterbody_area_m2'
        ]]

X_1.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 4665 entries, 0 to 4664
Data columns (total 16 columns):
 #   Column                    Non-Null Count  Dtype  
---  ------                    --------------  -----  
 0   log_area                  4665 non-null   float64
 1   Floor_clean               4665 non-null   int64  
 2   SaleType_encoded          4665 non-null   int64  
 3   bus_stop_count_1km        4665 non-null   int64  
 4   financial_facilities_1km  4665 non-null   int64  
 5   medical_facilities_1km    4665 non-null   int64  
 6   distance_to_metro         4665 non-null   float64
 7   distance_to_railway       4665 non-null   float64
 8   school_count_1km          4665 non-null   int64  
 9   kindergarten_count_1km    4665 non-null   int64  
 10  distance_to_supermarket   4665 non-null   float64
 11  distance_to_techhub       4665 non-null   float64
 12  distance_to_park          4665 non-null   float64
 13  distance_to_waterbody     4665 non-null   float64
 14  park_area_m2              4665 non-null   float64
 15  waterbody_area_m2         4665 non-null   float64
dtypes: float64(9), int64(7)
memory usage: 583.3 KB

from pysal.lib import weights
from pysal.explore import esda
import numpy
import pandas
import geopandas
import matplotlib.pyplot as plt
import seaborn
import contextily

C:\Users\Sayan\anaconda3\Lib\site-packages\spaghetti\network.py:41: FutureWarning: The next major release of pysal/spaghetti (2.0.0) will drop support for all ``libpysal.cg`` geometries. This change is a first step in refactoring ``spaghetti`` that is expected to result in dramatically reduced runtimes for network instantiation and operations. Users currently requiring network and point pattern input as ``libpysal.cg`` geometries should prepare for this simply by converting to ``shapely`` geometries.
  warnings.warn(dep_msg, FutureWarning, stacklevel=1)

import geopandas as gpd
from libpysal.weights import KNN
from shapely.geometry import Point

# Convert to GeoDataFrame
gdf = gpd.GeoDataFrame(
    df,
    geometry=gpd.points_from_xy(df['Longitude'], df['Latitude']),
    crs="EPSG:4326"
)

# Example: k-nearest neighbors (k=8 is common)
w = KNN.from_dataframe(gdf, k=8)
w.transform = "r"  # row-standardized

C:\Users\Sayan\anaconda3\Lib\site-packages\libpysal\weights\distance.py:153: UserWarning: The weights matrix is not fully connected: 
 There are 35 disconnected components.
  W.__init__(self, neighbors, id_order=ids, **kwargs)

from spreg import OLS, diagnostics_sp
from esda import Moran
import matplotlib.pyplot as plt

X_1

# Dependent variable
y = df['log_amount'].values.reshape(-1, 1)

# Run OLS (you've already defined y, X, and w)
ols_model = OLS(y, X, w=w, name_y='log_amount')

# Print full summary
print(ols_model.summary)

REGRESSION RESULTS
------------------

SUMMARY OF OUTPUT: ORDINARY LEAST SQUARES
-----------------------------------------
Data set            :     unknown
Weights matrix      :     unknown
Dependent Variable  :  log_amount                Number of Observations:        4665
Mean dependent var  :     15.7608                Number of Variables   :          26
S.D. dependent var  :      0.5348                Degrees of Freedom    :        4639
R-squared           :      0.2534
Adjusted R-squared  :      0.2494
Sum squared residual:     995.964                F-statistic           :     62.9959
Sigma-square        :       0.215                Prob(F-statistic)     :  1.966e-271
S.E. of regression  :       0.463                Log likelihood        :   -3017.661
Sigma-square ML     :       0.213                Akaike info criterion :    6087.321
S.E of regression ML:      0.4621                Schwarz criterion     :    6254.965

------------------------------------------------------------------------------------
            Variable     Coefficient       Std.Error     t-Statistic     Probability
------------------------------------------------------------------------------------
            CONSTANT        12.87346         0.16787        76.68839         0.00000
            log_area         0.49824         0.01743        28.58996         0.00000
         Floor_clean         0.00001         0.00000         2.08643         0.03699
    SaleType_encoded        -0.08521         0.01555        -5.47991         0.00000
  bus_stop_count_1km         0.00050         0.00133         0.37891         0.70477
financial_facilities_1km         0.00974         0.00124         7.83535         0.00000
medical_facilities_1km        -0.00529         0.00113        -4.67024         0.00000
   distance_to_metro        -0.00000         0.00000        -1.18401         0.23647
 distance_to_railway        -0.00003         0.00001        -4.74976         0.00000
    school_count_1km        -0.01003         0.00203        -4.94092         0.00000
kindergarten_count_1km         0.01079         0.00327         3.30409         0.00096
distance_to_supermarket        -0.00000         0.00002        -0.26503         0.79100
 distance_to_techhub        -0.00002         0.00001        -3.88549         0.00010
    distance_to_park        -0.00038         0.00004       -10.41730         0.00000
distance_to_waterbody        -0.00005         0.00002        -2.94350         0.00326
        park_area_m2         0.00000         0.00000         2.27834         0.02275
   waterbody_area_m2         0.00000         0.00000         1.69098         0.09091
 park_leisure_garden         0.08824         0.03424         2.57708         0.00999
   park_leisure_park        -0.04085         0.02277        -1.79403         0.07287
  park_leisure_pitch         0.01488         0.02293         0.64887         0.51646
park_leisure_playground        -0.06117         0.03054        -2.00312         0.04522
park_leisure_recreation_common         0.12988         0.07940         1.63587         0.10193
 waterbody_type_lake        -0.38751         0.11103        -3.49002         0.00049
 waterbody_type_pond        -0.30946         0.11157        -2.77361         0.00557
waterbody_type_reservoir        -0.32965         0.11306        -2.91580         0.00356
waterbody_type_river        -0.37704         0.11722        -3.21660         0.00131
------------------------------------------------------------------------------------

REGRESSION DIAGNOSTICS
MULTICOLLINEARITY CONDITION NUMBER          99.397

TEST ON NORMALITY OF ERRORS
TEST                             DF        VALUE           PROB
Jarque-Bera                       2        516.320           0.0000

DIAGNOSTICS FOR HETEROSKEDASTICITY
RANDOM COEFFICIENTS
TEST                             DF        VALUE           PROB
Breusch-Pagan test               25        985.222           0.0000
Koenker-Bassett test             25        558.714           0.0000
================================ END OF REPORT =====================================

# Run OLS (you've already defined y, X, and w)
ols_model_1 = OLS(y, X_1, w=w, name_y='log_amount')

# Print full summary
print(ols_model_1.summary)

REGRESSION RESULTS
------------------

SUMMARY OF OUTPUT: ORDINARY LEAST SQUARES
-----------------------------------------
Data set            :     unknown
Weights matrix      :     unknown
Dependent Variable  :  log_amount                Number of Observations:        4665
Mean dependent var  :     15.7608                Number of Variables   :          17
S.D. dependent var  :      0.5348                Degrees of Freedom    :        4648
R-squared           :      0.2428
Adjusted R-squared  :      0.2402
Sum squared residual:     1010.13                F-statistic           :     93.1645
Sigma-square        :       0.217                Prob(F-statistic)     :  6.541e-266
S.E. of regression  :       0.466                Log likelihood        :   -3050.606
Sigma-square ML     :       0.217                Akaike info criterion :    6135.212
S.E of regression ML:      0.4653                Schwarz criterion     :    6244.825

------------------------------------------------------------------------------------
            Variable     Coefficient       Std.Error     t-Statistic     Probability
------------------------------------------------------------------------------------
            CONSTANT        12.50966         0.13012        96.13973         0.00000
            log_area         0.49580         0.01752        28.30667         0.00000
         Floor_clean         0.00001         0.00000         2.18008         0.02930
    SaleType_encoded        -0.08561         0.01562        -5.48158         0.00000
  bus_stop_count_1km        -0.00038         0.00131        -0.28758         0.77368
financial_facilities_1km         0.01033         0.00123         8.36376         0.00000
medical_facilities_1km        -0.00479         0.00113        -4.23461         0.00002
   distance_to_metro         0.00000         0.00000         0.11785         0.90619
 distance_to_railway        -0.00003         0.00001        -4.81390         0.00000
    school_count_1km        -0.01042         0.00202        -5.15633         0.00000
kindergarten_count_1km         0.01087         0.00326         3.33820         0.00085
distance_to_supermarket         0.00000         0.00002         0.27484         0.78345
 distance_to_techhub        -0.00002         0.00001        -4.42819         0.00001
    distance_to_park        -0.00038         0.00004       -10.38167         0.00000
distance_to_waterbody        -0.00005         0.00002        -2.86311         0.00421
        park_area_m2         0.00000         0.00000         2.22295         0.02627
   waterbody_area_m2         0.00000         0.00000         0.42946         0.66761
------------------------------------------------------------------------------------

REGRESSION DIAGNOSTICS
MULTICOLLINEARITY CONDITION NUMBER          79.193

TEST ON NORMALITY OF ERRORS
TEST                             DF        VALUE           PROB
Jarque-Bera                       2        578.973           0.0000

DIAGNOSTICS FOR HETEROSKEDASTICITY
RANDOM COEFFICIENTS
TEST                             DF        VALUE           PROB
Breusch-Pagan test               16        997.241           0.0000
Koenker-Bassett test             16        552.442           0.0000
================================ END OF REPORT =====================================

'''
X_2 = X_1[['log_area', 'Floor_clean', 'SaleType_encoded',
        'bus_stop_count_1km', 'financial_facilities_1km',
        'medical_facilities_1km',
        'distance_to_metro', 'distance_to_railway', 'school_count_1km',
        'kindergarten_count_1km', 'distance_to_supermarket',
        'distance_to_techhub']]
'''

"\nX_2 = X_1[['log_area', 'Floor_clean', 'SaleType_encoded',\n        'bus_stop_count_1km', 'financial_facilities_1km',\n        'medical_facilities_1km',\n        'distance_to_metro', 'distance_to_railway', 'school_count_1km',\n        'kindergarten_count_1km', 'distance_to_supermarket',\n        'distance_to_techhub']]\n"

'''
# Run OLS (you've already defined y, X, and w)
ols_model_2 = OLS(y, X_2, w=w, name_y='log_amount')

# Print full summary
print(ols_model_2.summary)
'''

"\n# Run OLS (you've already defined y, X, and w)\nols_model_2 = OLS(y, X_2, w=w, name_y='log_amount')\n\n# Print full summary\nprint(ols_model_2.summary)\n"

from esda import Moran
import matplotlib.pyplot as plt

residuals = ols_model.u.flatten()  # from your fitted OLS model

mi = Moran(residuals, w)
print("Moran's I:", mi.I)
print("p-value:", mi.p_sim)

Moran's I: 0.2751378925229339
p-value: 0.001

from spreg import LMtests

# Assuming ols_model is your fitted OLS model and w is your spatial weights object
lm_tests = LMtests(ols_model, w)

# Accessing the results
print(f"LM Error: {lm_tests.lme[0]:.3f}, p-value: {lm_tests.lme[1]:.3f}")
print(f"LM Lag: {lm_tests.lml[0]:.3f}, p-value: {lm_tests.lml[1]:.3f}")
print(f"Robust LM Error: {lm_tests.rlme[0]:.3f}, p-value: {lm_tests.rlme[1]:.3f}")
print(f"Robust LM Lag: {lm_tests.rlml[0]:.3f}, p-value: {lm_tests.rlml[1]:.3f}")
print(f"LM SARMA: {lm_tests.sarma[0]:.3f}, p-value: {lm_tests.sarma[1]:.3f}")

LM Error: 1624.173, p-value: 0.000
LM Lag: 1394.095, p-value: 0.000
Robust LM Error: 241.437, p-value: 0.000
Robust LM Lag: 11.359, p-value: 0.001
LM SARMA: 1635.532, p-value: 0.000

import matplotlib.pyplot as plt
from splot.esda import moran_scatterplot

fig, ax = moran_scatterplot(mi, aspect_equal=True)
ax.set_title("Moran's I Scatterplot (OLS residuals)")
plt.show()

'''
import matplotlib.pyplot as plt
from splot.esda import moran_scatterplot
import os

# Create folder if it doesn't exist
os.makedirs("outputs", exist_ok=True)

# Plot Moran's I
fig, ax = moran_scatterplot(mi, aspect_equal=True)
ax.set_title("Moran's I Scatterplot (OLS residuals)")

# Save figure to 'outputs' folder
fig.savefig("outputs/moran_scatterplot.png", dpi=300, bbox_inches="tight")

plt.show()
'''

'\nimport matplotlib.pyplot as plt\nfrom splot.esda import moran_scatterplot\nimport os\n\n# Create folder if it doesn\'t exist\nos.makedirs("outputs", exist_ok=True)\n\n# Plot Moran\'s I\nfig, ax = moran_scatterplot(mi, aspect_equal=True)\nax.set_title("Moran\'s I Scatterplot (OLS residuals)")\n\n# Save figure to \'outputs\' folder\nfig.savefig("outputs/moran_scatterplot.png", dpi=300, bbox_inches="tight")\n\nplt.show()\n'

import matplotlib.pyplot as plt
from matplotlib.colors import ListedColormap
import matplotlib.patches as mpatches
import numpy as np

from esda import Moran_Local
import numpy as np

# y = variable you used for Moran's I (e.g., OLS residuals)
# w = spatial weights matrix (same as used for Moran's I)

lisa = Moran_Local(y, w)

gdf['local_I'] = lisa.Is
gdf['p_value'] = lisa.p_sim
gdf['quadrant'] = lisa.q
gdf['significant'] = lisa.p_sim < 0.05     # 5% significance

import matplotlib.pyplot as plt
import matplotlib.colors as colors

# Only clusters that are statistically significant
sig = gdf['significant']

# Map cluster categories
cluster_labels = {
    1: "High-High",
    2: "Low-High",
    3: "Low-Low",
    4: "High-Low"
}

# Create a color map
cluster_colors = {
    "High-High": "red",
    "Low-Low": "blue",
    "High-Low": "lightpink",
    "Low-High": "lightblue"
}

gdf['cluster_type'] = gdf.apply(
    lambda row: cluster_labels[row['quadrant']] if row['significant'] else "Not Significant",
    axis=1
)

# Plot
fig, ax = plt.subplots(figsize=(10, 10))
gdf.plot(column='cluster_type',
         cmap=colors.ListedColormap(["red","blue","lightpink","lightblue","lightgrey"]),
         legend=True,
         ax=ax)

ax.set_title("LISA Cluster Map (Local Moran’s I)", fontsize=16)
ax.axis("off")

plt.show()

import geopandas as gpd
import matplotlib.pyplot as plt
import matplotlib.colors as colors

# ----------------------------------------------------------
# 1. LOAD DATA
# ----------------------------------------------------------
# Your housing GDF (already includes LISA results)
# gdf = ...

# Load BBMP boundary shapefile
bbmp = gpd.read_file("BBMP_shape.shp")

# ----------------------------------------------------------
# 2. PROJECT BOTH LAYERS (recommended: Web Mercator 3857)
# ----------------------------------------------------------
gdf_web = gdf.to_crs(epsg=3857)
bbmp_web = bbmp.to_crs(epsg=3857)

# ----------------------------------------------------------
# 3. MAP LISA CLUSTERS (color-blind friendly palette)
# ----------------------------------------------------------
cluster_labels = {
    1: "High-High",
    2: "Low-High",
    3: "Low-Low",
    4: "High-Low"
}

# Add cluster type with significance filter
gdf_web["cluster_type"] = gdf_web.apply(
    lambda row: cluster_labels[row["quadrant"]] if row["significant"] else "Not Significant",
    axis=1
)

# Colour-blind friendly palette ("Tol palette")
cluster_palette = {
    "High-High": "#D81B60",      # reddish-pink
    "Low-Low": "#1E88E5",        # strong blue
    "High-Low": "#FFC107",       # yellow
    "Low-High": "#004D40",       # dark teal
    "Not Significant": "#BDBDBD" # grey
}

# Manual colormap
cmap = colors.ListedColormap(list(cluster_palette.values()))
norm = colors.ListedColormap

# ----------------------------------------------------------
# 4. CREATE FIGURE
# ----------------------------------------------------------
fig, ax = plt.subplots(figsize=(12, 12))

# Plot BBMP boundary first
bbmp_web.boundary.plot(ax=ax, color="black", linewidth=0.5)

# Plot LISA clusters
gdf_web.plot(
    column="cluster_type",
    ax=ax,
    categorical=True,
    legend=True,
    linewidth=0.2,
    edgecolor="white",
    cmap=colors.ListedColormap([cluster_palette[k] for k in cluster_palette.keys()])
)

# ----------------------------------------------------------
# 5. STYLING
# ----------------------------------------------------------
ax.set_title("LISA Cluster Map – BBMP, Bengaluru (Local Moran’s I)", fontsize=18)
ax.axis("off")

plt.show()

import geopandas as gpd
import matplotlib.pyplot as plt
import matplotlib.colors as colors
import contextily as ctx

# ----------------------------------------------------------
# 1. LOAD DATA
# ----------------------------------------------------------
# Your spatial data frame with LISA results:
# gdf = ...

# BBMP boundary
bbmp = gpd.read_file("BBMP_shape.shp")

# ----------------------------------------------------------
# 2. PROJECT EVERYTHING TO WEB MERCATOR 3857
# ----------------------------------------------------------
gdf_web = gdf.to_crs(epsg=3857)
bbmp_web = bbmp.to_crs(epsg=3857)

# ----------------------------------------------------------
# 3. CLASSIFY CLUSTERS
# ----------------------------------------------------------
cluster_labels = {
    1: "High-High",
    2: "Low-High",
    3: "Low-Low",
    4: "High-Low"
}

# Assign cluster types only for significant observations
gdf_web["cluster_type"] = gdf_web.apply(
    lambda row: cluster_labels[row["quadrant"]] if row["significant"] else "Not Significant",
    axis=1
)

# Colour-blind friendly palette
cluster_palette = {
    "High-High": "#D81B60",      # reddish-pink
    "Low-Low": "#1E88E5",        # blue
    "High-Low": "#FFC107",       # yellow
    "Low-High": "#004D40",       # dark teal
    "Not Significant": "#BDBDBD" # neutral grey
}

# ----------------------------------------------------------
# 4. PLOT FIGURE
# ----------------------------------------------------------
fig, ax = plt.subplots(figsize=(13, 13))

# 4A. Add basemap FIRST
# (Empty plot so basemap draws below everything)
bbmp_web.to_crs(epsg=3857).plot(ax=ax, alpha=0)  # invisible placeholder to set extent

ctx.add_basemap(
    ax,
    source=ctx.providers.CartoDB.Positron,
    zoom=12
)

# 4B. Plot BBMP boundary
bbmp_web.boundary.plot(ax=ax, color="black", linewidth=0.7)

# 4C. Plot LISA clusters
gdf_web.plot(
    column="cluster_type",
    ax=ax,
    legend=True,
    categorical=True,
    linewidth=0.3,
    edgecolor="white",
    cmap=colors.ListedColormap([cluster_palette[c] for c in cluster_palette])
)

# ----------------------------------------------------------
# 5. STYLING
# ----------------------------------------------------------
ax.set_title("LISA Cluster Map – BBMP, Bengaluru (Local Moran’s I)", fontsize=18)
ax.axis("off")

# ----------------------------------------------------------
# 6. SAVE HIGH-RES OUTPUT
# ----------------------------------------------------------
output_path = "outputs/lisa_cluster_bbmp_osm.png"
plt.savefig(output_path, dpi=600, bbox_inches="tight")
plt.close()

print(f"✅ LISA map with OSM basemap saved → {output_path}")

✅ LISA map with OSM basemap saved → outputs/lisa_cluster_bbmp_osm.png

from spreg import ML_Error

# --- Spatial Error Model ---
sem_model = ML_Error(y, X, w=w, name_y='log_amount')

# --- Summary ---
print(sem_model.summary)

C:\Users\Sayan\anaconda3\Lib\site-packages\spreg\ml_error.py:184: RuntimeWarning: Method 'bounded' does not support relative tolerance in x; defaulting to absolute tolerance.
  res = minimize_scalar(

REGRESSION RESULTS
------------------

SUMMARY OF OUTPUT: ML SPATIAL ERROR (METHOD = full)
---------------------------------------------------
Data set            :     unknown
Weights matrix      :     unknown
Dependent Variable  :  log_amount                Number of Observations:        4665
Mean dependent var  :     15.7608                Number of Variables   :          26
S.D. dependent var  :      0.5348                Degrees of Freedom    :        4639
Pseudo R-squared    :      0.2463
Log likelihood      :  -2586.5186
Sigma-square ML     :      0.1695                Akaike info criterion :    5225.037
S.E of regression   :      0.4117                Schwarz criterion     :    5392.681

------------------------------------------------------------------------------------
            Variable     Coefficient       Std.Error     z-Statistic     Probability
------------------------------------------------------------------------------------
            CONSTANT        13.06960         0.22321        58.55260         0.00000
            log_area         0.48499         0.01651        29.37162         0.00000
         Floor_clean         0.00001         0.00000         1.76428         0.07768
    SaleType_encoded        -0.08610         0.01529        -5.63025         0.00000
  bus_stop_count_1km         0.00129         0.00226         0.57039         0.56841
financial_facilities_1km         0.00596         0.00199         2.99286         0.00276
medical_facilities_1km        -0.00386         0.00198        -1.94539         0.05173
   distance_to_metro        -0.00000         0.00001        -0.23950         0.81072
 distance_to_railway        -0.00002         0.00001        -1.56572         0.11741
    school_count_1km        -0.00321         0.00340        -0.94557         0.34437
kindergarten_count_1km         0.00702         0.00614         1.14291         0.25307
distance_to_supermarket        -0.00004         0.00003        -1.21798         0.22323
 distance_to_techhub        -0.00002         0.00001        -1.87366         0.06098
    distance_to_park        -0.00036         0.00005        -7.00329         0.00000
distance_to_waterbody        -0.00003         0.00003        -1.08235         0.27910
        park_area_m2         0.00000         0.00000         1.14979         0.25023
   waterbody_area_m2         0.00000         0.00000         1.51846         0.12890
 park_leisure_garden         0.05015         0.04084         1.22803         0.21944
   park_leisure_park        -0.02732         0.02790        -0.97905         0.32756
  park_leisure_pitch         0.01997         0.02781         0.71816         0.47266
park_leisure_playground        -0.03781         0.03727        -1.01437         0.31041
park_leisure_recreation_common         0.03083         0.09291         0.33185         0.74000
 waterbody_type_lake        -0.55142         0.18569        -2.96958         0.00298
 waterbody_type_pond        -0.49897         0.18555        -2.68921         0.00716
waterbody_type_reservoir        -0.50966         0.18848        -2.70410         0.00685
waterbody_type_river        -0.52593         0.19341        -2.71931         0.00654
              lambda         0.58059         0.01605        36.17973         0.00000
------------------------------------------------------------------------------------
================================ END OF REPORT =====================================

from spreg import ML_Error

# --- Spatial Error Model ---
sem_model_1 = ML_Error(y, X_1, w=w, name_y='log_amount')

# --- Summary ---
print(sem_model_1.summary)

C:\Users\Sayan\anaconda3\Lib\site-packages\spreg\ml_error.py:184: RuntimeWarning: Method 'bounded' does not support relative tolerance in x; defaulting to absolute tolerance.
  res = minimize_scalar(

REGRESSION RESULTS
------------------

SUMMARY OF OUTPUT: ML SPATIAL ERROR (METHOD = full)
---------------------------------------------------
Data set            :     unknown
Weights matrix      :     unknown
Dependent Variable  :  log_amount                Number of Observations:        4665
Mean dependent var  :     15.7608                Number of Variables   :          17
S.D. dependent var  :      0.5348                Degrees of Freedom    :        4648
Pseudo R-squared    :      0.2367
Log likelihood      :  -2597.6105
Sigma-square ML     :      0.1700                Akaike info criterion :    5229.221
S.E of regression   :      0.4124                Schwarz criterion     :    5338.834

------------------------------------------------------------------------------------
            Variable     Coefficient       Std.Error     z-Statistic     Probability
------------------------------------------------------------------------------------
            CONSTANT        12.54911         0.13235        94.81613         0.00000
            log_area         0.48256         0.01652        29.21551         0.00000
         Floor_clean         0.00001         0.00000         1.80679         0.07080
    SaleType_encoded        -0.08790         0.01530        -5.74553         0.00000
  bus_stop_count_1km         0.00094         0.00227         0.41497         0.67816
financial_facilities_1km         0.00620         0.00200         3.10435         0.00191
medical_facilities_1km        -0.00390         0.00199        -1.95426         0.05067
   distance_to_metro         0.00000         0.00001         0.11247         0.91045
 distance_to_railway        -0.00002         0.00001        -1.37226         0.16998
    school_count_1km        -0.00323         0.00343        -0.94216         0.34611
kindergarten_count_1km         0.00700         0.00620         1.12899         0.25890
distance_to_supermarket        -0.00003         0.00003        -1.14969         0.25027
 distance_to_techhub        -0.00002         0.00001        -2.16373         0.03049
    distance_to_park        -0.00036         0.00005        -7.00452         0.00000
distance_to_waterbody        -0.00003         0.00003        -1.05035         0.29356
        park_area_m2         0.00000         0.00000         1.20651         0.22762
   waterbody_area_m2         0.00000         0.00000         1.02931         0.30333
              lambda         0.58748         0.01583        37.12110         0.00000
------------------------------------------------------------------------------------
================================ END OF REPORT =====================================

from spreg import ML_Lag

# --- Spatial Lag Model ---
slm_model = ML_Lag(y, X, w=w, name_y='log_amount')

# --- Summary ---
print(slm_model.summary)

REGRESSION RESULTS
------------------

SUMMARY OF OUTPUT: MAXIMUM LIKELIHOOD SPATIAL LAG (METHOD = FULL)
-----------------------------------------------------------------
Data set            :     unknown
Weights matrix      :     unknown
Dependent Variable  :  log_amount                Number of Observations:        4665
Mean dependent var  :     15.7608                Number of Variables   :          27
S.D. dependent var  :      0.5348                Degrees of Freedom    :        4638
Pseudo R-squared    :      0.3947
Spatial Pseudo R-squared:  0.2422
Log likelihood      :  -2619.6914
Sigma-square ML     :      0.1737                Akaike info criterion :    5293.383
S.E of regression   :      0.4168                Schwarz criterion     :    5467.475

------------------------------------------------------------------------------------
            Variable     Coefficient       Std.Error     z-Statistic     Probability
------------------------------------------------------------------------------------
            CONSTANT         4.73382         0.28530        16.59252         0.00000
            log_area         0.44044         0.01592        27.65873         0.00000
         Floor_clean         0.00001         0.00000         1.78350         0.07451
    SaleType_encoded        -0.06951         0.01401        -4.96235         0.00000
  bus_stop_count_1km        -0.00082         0.00120        -0.68796         0.49148
financial_facilities_1km         0.00449         0.00113         3.98336         0.00007
medical_facilities_1km        -0.00168         0.00102        -1.64525         0.09992
   distance_to_metro        -0.00000         0.00000        -1.25982         0.20773
 distance_to_railway        -0.00001         0.00001        -2.36374         0.01809
    school_count_1km        -0.00459         0.00183        -2.50621         0.01220
kindergarten_count_1km         0.00292         0.00294         0.99192         0.32124
distance_to_supermarket         0.00001         0.00002         0.60447         0.54553
 distance_to_techhub        -0.00001         0.00000        -1.41252         0.15780
    distance_to_park        -0.00022         0.00003        -6.54467         0.00000
distance_to_waterbody        -0.00003         0.00002        -1.65230         0.09847
        park_area_m2         0.00000         0.00000         0.38296         0.70175
   waterbody_area_m2         0.00000         0.00000         1.43595         0.15102
 park_leisure_garden         0.05183         0.03080         1.68282         0.09241
   park_leisure_park        -0.02993         0.02049        -1.46074         0.14409
  park_leisure_pitch         0.00732         0.02063         0.35471         0.72281
park_leisure_playground        -0.04145         0.02748        -1.50819         0.13151
park_leisure_recreation_common         0.03288         0.07143         0.46029         0.64531
 waterbody_type_lake        -0.23071         0.10011        -2.30458         0.02119
 waterbody_type_pond        -0.19092         0.10054        -1.89887         0.05758
waterbody_type_reservoir        -0.18988         0.10186        -1.86406         0.06231
waterbody_type_river        -0.21463         0.10561        -2.03227         0.04213
        W_log_amount         0.52432         0.01594        32.90025         0.00000
------------------------------------------------------------------------------------

SPATIAL LAG MODEL IMPACTS
Impacts computed using the 'simple' method.
            Variable         Direct        Indirect          Total
            log_area         0.4404          0.4855          0.9259
         Floor_clean         0.0000          0.0000          0.0000
    SaleType_encoded        -0.0695         -0.0766         -0.1461
  bus_stop_count_1km        -0.0008         -0.0009         -0.0017
financial_facilities_1km         0.0045          0.0049          0.0094
medical_facilities_1km        -0.0017         -0.0019         -0.0035
   distance_to_metro        -0.0000         -0.0000         -0.0000
 distance_to_railway        -0.0000         -0.0000         -0.0000
    school_count_1km        -0.0046         -0.0051         -0.0096
kindergarten_count_1km         0.0029          0.0032          0.0061
distance_to_supermarket         0.0000          0.0000          0.0000
 distance_to_techhub        -0.0000         -0.0000         -0.0000
    distance_to_park        -0.0002         -0.0002         -0.0005
distance_to_waterbody        -0.0000         -0.0000         -0.0001
        park_area_m2         0.0000          0.0000          0.0000
   waterbody_area_m2         0.0000          0.0000          0.0000
 park_leisure_garden         0.0518          0.0571          0.1090
   park_leisure_park        -0.0299         -0.0330         -0.0629
  park_leisure_pitch         0.0073          0.0081          0.0154
park_leisure_playground        -0.0415         -0.0457         -0.0871
park_leisure_recreation_common         0.0329          0.0362          0.0691
 waterbody_type_lake        -0.2307         -0.2543         -0.4850
 waterbody_type_pond        -0.1909         -0.2104         -0.4014
waterbody_type_reservoir        -0.1899         -0.2093         -0.3992
waterbody_type_river        -0.2146         -0.2366         -0.4512
================================ END OF REPORT =====================================

from spreg import ML_Lag

# --- Spatial Lag Model ---
slm_model_1 = ML_Lag(y, X_1, w=w, name_y='log_amount')

# --- Summary ---
print(slm_model_1.summary)

REGRESSION RESULTS
------------------

SUMMARY OF OUTPUT: MAXIMUM LIKELIHOOD SPATIAL LAG (METHOD = FULL)
-----------------------------------------------------------------
Data set            :     unknown
Weights matrix      :     unknown
Dependent Variable  :  log_amount                Number of Observations:        4665
Mean dependent var  :     15.7608                Number of Variables   :          18
S.D. dependent var  :      0.5348                Degrees of Freedom    :        4647
Pseudo R-squared    :      0.3922
Spatial Pseudo R-squared:  0.2304
Log likelihood      :  -2633.8689
Sigma-square ML     :      0.1745                Akaike info criterion :    5303.738
S.E of regression   :      0.4177                Schwarz criterion     :    5419.799

------------------------------------------------------------------------------------
            Variable     Coefficient       Std.Error     z-Statistic     Probability
------------------------------------------------------------------------------------
            CONSTANT         4.37808         0.26032        16.81826         0.00000
            log_area         0.43811         0.01594        27.49324         0.00000
         Floor_clean         0.00001         0.00000         1.84520         0.06501
    SaleType_encoded        -0.06972         0.01402        -4.97357         0.00000
  bus_stop_count_1km        -0.00149         0.00117        -1.27244         0.20322
financial_facilities_1km         0.00477         0.00111         4.28054         0.00002
medical_facilities_1km        -0.00134         0.00102        -1.32133         0.18639
   distance_to_metro        -0.00000         0.00000        -0.47065         0.63789
 distance_to_railway        -0.00001         0.00001        -2.40006         0.01639
    school_count_1km        -0.00483         0.00181        -2.66346         0.00773
kindergarten_count_1km         0.00284         0.00292         0.97097         0.33156
distance_to_supermarket         0.00002         0.00002         1.01730         0.30901
 distance_to_techhub        -0.00001         0.00000        -1.70910         0.08743
    distance_to_park        -0.00021         0.00003        -6.39919         0.00000
distance_to_waterbody        -0.00003         0.00002        -1.55840         0.11914
        park_area_m2         0.00000         0.00000         0.32814         0.74281
   waterbody_area_m2         0.00000         0.00000         0.67907         0.49710
        W_log_amount         0.53324         0.01576        33.84253         0.00000
------------------------------------------------------------------------------------

SPATIAL LAG MODEL IMPACTS
Impacts computed using the 'simple' method.
            Variable         Direct        Indirect          Total
            log_area         0.4381          0.5005          0.9386
         Floor_clean         0.0000          0.0000          0.0000
    SaleType_encoded        -0.0697         -0.0796         -0.1494
  bus_stop_count_1km        -0.0015         -0.0017         -0.0032
financial_facilities_1km         0.0048          0.0054          0.0102
medical_facilities_1km        -0.0013         -0.0015         -0.0029
   distance_to_metro        -0.0000         -0.0000         -0.0000
 distance_to_railway        -0.0000         -0.0000         -0.0000
    school_count_1km        -0.0048         -0.0055         -0.0104
kindergarten_count_1km         0.0028          0.0032          0.0061
distance_to_supermarket         0.0000          0.0000          0.0000
 distance_to_techhub        -0.0000         -0.0000         -0.0000
    distance_to_park        -0.0002         -0.0002         -0.0004
distance_to_waterbody        -0.0000         -0.0000         -0.0001
        park_area_m2         0.0000          0.0000          0.0000
   waterbody_area_m2         0.0000          0.0000          0.0000
================================ END OF REPORT =====================================

from spreg import GM_Combo_Het

# --- SARAR (GMM version) ---
sarar_gmm = GM_Combo_Het(
    y,
    X,
    w=w,
    name_y='log_amount'
)

print(sarar_gmm.summary)

REGRESSION RESULTS
------------------

SUMMARY OF OUTPUT: SPATIALLY WEIGHTED 2SLS- GM-COMBO MODEL (HET)
----------------------------------------------------------------
Data set            :     unknown
Weights matrix      :     unknown
Dependent Variable  :  log_amount                Number of Observations:        4665
Mean dependent var  :     15.7608                Number of Variables   :          27
S.D. dependent var  :      0.5348                Degrees of Freedom    :        4638
Pseudo R-squared    :      0.3339
Spatial Pseudo R-squared:  0.2502
N. of iterations    :           1                Step1c computed       :          No

------------------------------------------------------------------------------------
            Variable     Coefficient       Std.Error     z-Statistic     Probability
------------------------------------------------------------------------------------
            CONSTANT         9.50360         1.41938         6.69561         0.00000
            log_area         0.48378         0.03522        13.73422         0.00000
         Floor_clean         0.00001         0.00000         8.98407         0.00000
    SaleType_encoded        -0.08417         0.01604        -5.24836         0.00000
  bus_stop_count_1km         0.00022         0.00216         0.10112         0.91945
financial_facilities_1km         0.00571         0.00209         2.73861         0.00617
medical_facilities_1km        -0.00310         0.00184        -1.68353         0.09227
   distance_to_metro        -0.00000         0.00000        -0.47519         0.63465
 distance_to_railway        -0.00002         0.00001        -1.61853         0.10555
    school_count_1km        -0.00426         0.00295        -1.44683         0.14795
kindergarten_count_1km         0.00537         0.00546         0.98251         0.32585
distance_to_supermarket        -0.00002         0.00003        -0.63886         0.52292
 distance_to_techhub        -0.00001         0.00001        -1.40858         0.15896
    distance_to_park        -0.00031         0.00005        -6.12755         0.00000
distance_to_waterbody        -0.00004         0.00003        -1.24871         0.21177
        park_area_m2         0.00000         0.00000         1.47241         0.14091
   waterbody_area_m2         0.00000         0.00000         1.30810         0.19084
 park_leisure_garden         0.05579         0.03854         1.44755         0.14774
   park_leisure_park        -0.02822         0.02750        -1.02614         0.30482
  park_leisure_pitch         0.01645         0.02625         0.62676         0.53082
park_leisure_playground        -0.04163         0.03440        -1.21023         0.22619
park_leisure_recreation_common         0.03133         0.11042         0.28374         0.77661
 waterbody_type_lake        -0.44360         0.25206        -1.75988         0.07843
 waterbody_type_pond        -0.39358         0.25207        -1.56137         0.11844
waterbody_type_reservoir        -0.40034         0.25376        -1.57760         0.11466
waterbody_type_river        -0.41660         0.25807        -1.61427         0.10647
        W_log_amount         0.21869         0.08171         2.67626         0.00744
              lambda         0.48136         0.06665         7.22237         0.00000
------------------------------------------------------------------------------------
Instrumented: W_log_amount
Instruments: W_Floor_clean, W_SaleType_encoded, W_bus_stop_count_1km,
             W_distance_to_metro, W_distance_to_park, W_distance_to_railway,
             W_distance_to_supermarket, W_distance_to_techhub,
             W_distance_to_waterbody, W_financial_facilities_1km,
             W_kindergarten_count_1km, W_log_area, W_medical_facilities_1km,
             W_park_area_m2, W_park_leisure_garden, W_park_leisure_park,
             W_park_leisure_pitch, W_park_leisure_playground,
             W_park_leisure_recreation_common, W_school_count_1km,
             W_waterbody_area_m2, W_waterbody_type_lake,
             W_waterbody_type_pond, W_waterbody_type_reservoir,
             W_waterbody_type_river
================================ END OF REPORT =====================================

import numpy as np

# number of observations
n = sarar_gmm.n  # or len(y)

# number of parameters estimated (intercept + betas + spatial lags)
k = sarar_gmm.k

# residuals
residuals = sarar_gmm.u  # residuals vector

# sum of squared residuals
ssr = np.sum(residuals ** 2)

# pseudo log-likelihood under Gaussian errors
llf = -n/2 * (np.log(2 * np.pi) + 1 + np.log(ssr / n))

# approximate AIC
AIC = -2 * llf + 2 * k

print(f"Approximate AIC (GMM-SARAR): {AIC:.3f}")

Approximate AIC (GMM-SARAR): 5584.546

from esda.moran import Moran
import numpy as np

# Moran's I for residuals
mi_ols  = Moran(ols_model.u, w)
mi_slm  = Moran(slm_model.u, w)
mi_sem  = Moran(sem_model.u, w)
mi_sarar = Moran(sarar_gmm.u, w)

print("Moran's I of residuals:")
print(f"OLS:   {mi_ols.I:.3f}")
print(f"SLM:   {mi_slm.I:.3f}")
print(f"SEM:   {mi_sem.I:.3f}")
print(f"SARAR: {mi_sarar.I:.3f}")

Moran's I of residuals:
OLS:   0.275
SLM:   0.037
SEM:   0.288
SARAR: 0.175

pip install mgwr

Requirement already satisfied: mgwr in c:\users\sayan\anaconda3\lib\site-packages (2.2.1)
Requirement already satisfied: scipy>=0.11 in c:\users\sayan\anaconda3\lib\site-packages (from mgwr) (1.13.1)
Requirement already satisfied: numpy>=1.3 in c:\users\sayan\anaconda3\lib\site-packages (from mgwr) (1.26.4)
Requirement already satisfied: libpysal>=4.0.0 in c:\users\sayan\anaconda3\lib\site-packages (from mgwr) (4.13.0)
Requirement already satisfied: spglm>=1.0.6 in c:\users\sayan\anaconda3\lib\site-packages (from mgwr) (1.1.0)
Requirement already satisfied: spreg in c:\users\sayan\anaconda3\lib\site-packages (from mgwr) (1.8.3)
Requirement already satisfied: beautifulsoup4>=4.10 in c:\users\sayan\anaconda3\lib\site-packages (from libpysal>=4.0.0->mgwr) (4.12.3)
Requirement already satisfied: geopandas>=0.10.0 in c:\users\sayan\anaconda3\lib\site-packages (from libpysal>=4.0.0->mgwr) (1.0.1)
Requirement already satisfied: packaging>=22 in c:\users\sayan\anaconda3\lib\site-packages (from libpysal>=4.0.0->mgwr) (24.1)
Requirement already satisfied: pandas>=1.4 in c:\users\sayan\anaconda3\lib\site-packages (from libpysal>=4.0.0->mgwr) (2.2.2)
Requirement already satisfied: platformdirs>=2.0.2 in c:\users\sayan\anaconda3\lib\site-packages (from libpysal>=4.0.0->mgwr) (3.10.0)
Requirement already satisfied: requests>=2.27 in c:\users\sayan\anaconda3\lib\site-packages (from libpysal>=4.0.0->mgwr) (2.32.3)
Requirement already satisfied: shapely>=2.0.1 in c:\users\sayan\anaconda3\lib\site-packages (from libpysal>=4.0.0->mgwr) (2.0.7)
Requirement already satisfied: scikit-learn>=1.1 in c:\users\sayan\anaconda3\lib\site-packages (from libpysal>=4.0.0->mgwr) (1.5.1)
Requirement already satisfied: soupsieve>1.2 in c:\users\sayan\anaconda3\lib\site-packages (from beautifulsoup4>=4.10->libpysal>=4.0.0->mgwr) (2.5)
Requirement already satisfied: pyogrio>=0.7.2 in c:\users\sayan\anaconda3\lib\site-packages (from geopandas>=0.10.0->libpysal>=4.0.0->mgwr) (0.10.0)
Requirement already satisfied: pyproj>=3.3.0 in c:\users\sayan\anaconda3\lib\site-packages (from geopandas>=0.10.0->libpysal>=4.0.0->mgwr) (3.7.1)
Requirement already satisfied: python-dateutil>=2.8.2 in c:\users\sayan\anaconda3\lib\site-packages (from pandas>=1.4->libpysal>=4.0.0->mgwr) (2.9.0.post0)
Requirement already satisfied: pytz>=2020.1 in c:\users\sayan\anaconda3\lib\site-packages (from pandas>=1.4->libpysal>=4.0.0->mgwr) (2024.1)
Requirement already satisfied: tzdata>=2022.7 in c:\users\sayan\anaconda3\lib\site-packages (from pandas>=1.4->libpysal>=4.0.0->mgwr) (2023.3)
Requirement already satisfied: charset-normalizer<4,>=2 in c:\users\sayan\anaconda3\lib\site-packages (from requests>=2.27->libpysal>=4.0.0->mgwr) (3.3.2)
Requirement already satisfied: idna<4,>=2.5 in c:\users\sayan\anaconda3\lib\site-packages (from requests>=2.27->libpysal>=4.0.0->mgwr) (3.7)
Requirement already satisfied: urllib3<3,>=1.21.1 in c:\users\sayan\anaconda3\lib\site-packages (from requests>=2.27->libpysal>=4.0.0->mgwr) (2.2.3)
Requirement already satisfied: certifi>=2017.4.17 in c:\users\sayan\anaconda3\lib\site-packages (from requests>=2.27->libpysal>=4.0.0->mgwr) (2025.1.31)
Requirement already satisfied: joblib>=1.2.0 in c:\users\sayan\anaconda3\lib\site-packages (from scikit-learn>=1.1->libpysal>=4.0.0->mgwr) (1.4.2)
Requirement already satisfied: threadpoolctl>=3.1.0 in c:\users\sayan\anaconda3\lib\site-packages (from scikit-learn>=1.1->libpysal>=4.0.0->mgwr) (3.5.0)
Requirement already satisfied: six>=1.5 in c:\users\sayan\anaconda3\lib\site-packages (from python-dateutil>=2.8.2->pandas>=1.4->libpysal>=4.0.0->mgwr) (1.16.0)
Note: you may need to restart the kernel to use updated packages.

# Find all boolean columns
bool_cols = df.select_dtypes(include=['bool']).columns
print(bool_cols)
# Convert all boolean columns to integers
df[bool_cols] = df[bool_cols].astype(int)

Index(['park_leisure_garden', 'park_leisure_park', 'park_leisure_pitch',
       'park_leisure_playground', 'park_leisure_recreation_common',
       'waterbody_type_lake', 'waterbody_type_pond',
       'waterbody_type_reservoir', 'waterbody_type_river'],
      dtype='object')

df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 4665 entries, 0 to 4664
Data columns (total 58 columns):
 #   Column                          Non-Null Count  Dtype  
---  ------                          --------------  -----  
 0   Area (in Sq. ft)                4665 non-null   float64
 1   Amount                          4665 non-null   float64
 2   Market Value                    4665 non-null   float64
 3   City                            4665 non-null   object 
 4   Project Name                    4665 non-null   object 
 5   Location                        4665 non-null   object 
 6   Latitude                        4665 non-null   float64
 7   Longitude                       4665 non-null   float64
 8   Sale Type                       4665 non-null   object 
 9   Registration Date               4665 non-null   object 
 10  Quaters                         4665 non-null   object 
 11  YR                              4665 non-null   int64  
 12  Amount in CR                    4665 non-null   float64
 13  Price Range                     4665 non-null   object 
 14  Project Name main               2306 non-null   object 
 15  Rate (Rs/sft)                   4665 non-null   object 
 16  Unit                            4301 non-null   object 
 17  Floor                           4506 non-null   object 
 18  Tower                           962 non-null    object 
 19  Wing                            1438 non-null   object 
 20  Party 1                         4665 non-null   object 
 21  Party 2                         4665 non-null   object 
 22  Count                           4665 non-null   object 
 23  bus_stop_count_1km              4665 non-null   int64  
 24  financial_facilities_1km        4665 non-null   int64  
 25  medical_facilities_1km          4665 non-null   int64  
 26  hospital_facilities_1km         4665 non-null   int64  
 27  distance_to_metro               4665 non-null   float64
 28  distance_to_railway             4665 non-null   float64
 29  school_count_1km                4665 non-null   int64  
 30  kindergarten_count_1km          4665 non-null   int64  
 31  distance_to_supermarket         4665 non-null   float64
 32  distance_to_techhub             4665 non-null   float64
 33  distance_to_park                4665 non-null   float64
 34  distance_to_waterbody           4665 non-null   float64
 35  park_name                       750 non-null    object 
 36  park_area_m2                    4665 non-null   float64
 37  park_lat                        4665 non-null   float64
 38  park_lon                        4665 non-null   float64
 39  waterbody_name                  3080 non-null   object 
 40  waterbody_waterbody_type_clean  4665 non-null   object 
 41  waterbody_area_m2               4665 non-null   float64
 42  waterbody_lat                   4665 non-null   float64
 43  waterbody_lon                   4665 non-null   float64
 44  Floor_clean                     4665 non-null   int64  
 45  SaleType_encoded                4665 non-null   int64  
 46  park_leisure_garden             4665 non-null   int32  
 47  park_leisure_park               4665 non-null   int32  
 48  park_leisure_pitch              4665 non-null   int32  
 49  park_leisure_playground         4665 non-null   int32  
 50  park_leisure_recreation_common  4665 non-null   int32  
 51  waterbody_type_lake             4665 non-null   int32  
 52  waterbody_type_pond             4665 non-null   int32  
 53  waterbody_type_reservoir        4665 non-null   int32  
 54  waterbody_type_river            4665 non-null   int32  
 55  log_amount                      4665 non-null   float64
 56  area_raw                        4665 non-null   float64
 57  log_area                        4665 non-null   float64
dtypes: float64(21), int32(9), int64(9), object(19)
memory usage: 1.9+ MB

df.columns

Index(['Area (in Sq. ft)', 'Amount', 'Market Value', 'City', 'Project Name',
       'Location', 'Latitude', 'Longitude', 'Sale Type', 'Registration Date',
       'Quaters', 'YR', 'Amount in CR', 'Price Range', 'Project Name main',
       'Rate (Rs/sft)', 'Unit', 'Floor', 'Tower', 'Wing', 'Party 1', 'Party 2',
       'Count', 'bus_stop_count_1km', 'financial_facilities_1km',
       'medical_facilities_1km', 'hospital_facilities_1km',
       'distance_to_metro', 'distance_to_railway', 'school_count_1km',
       'kindergarten_count_1km', 'distance_to_supermarket',
       'distance_to_techhub', 'distance_to_park', 'distance_to_waterbody',
       'park_name', 'park_area_m2', 'park_lat', 'park_lon', 'waterbody_name',
       'waterbody_waterbody_type_clean', 'waterbody_area_m2', 'waterbody_lat',
       'waterbody_lon', 'Floor_clean', 'SaleType_encoded',
       'park_leisure_garden', 'park_leisure_park', 'park_leisure_pitch',
       'park_leisure_playground', 'park_leisure_recreation_common',
       'waterbody_type_lake', 'waterbody_type_pond',
       'waterbody_type_reservoir', 'waterbody_type_river', 'log_amount',
       'area_raw', 'log_area'],
      dtype='object')

# Drop 
df_dropped = df.drop(['Area (in Sq. ft)', 'Amount', 'Market Value', 'City', 'Project Name',
       'Location', 'Sale Type', 'Registration Date','hospital_facilities_1km',
       'Quaters', 'YR', 'Amount in CR', 'Price Range', 'Project Name main',
       'Rate (Rs/sft)', 'Unit', 'Floor', 'Tower', 'Wing', 'Party 1', 'Party 2',
       'Count','park_lat', 'park_lon', 'waterbody_name','park_name','waterbody_waterbody_type_clean', 'waterbody_lat',
       'waterbody_lon',], axis=1) # axis=1 specifies that we are dropping a column

df_dropped.columns

Index(['Latitude', 'Longitude', 'bus_stop_count_1km',
       'financial_facilities_1km', 'medical_facilities_1km',
       'distance_to_metro', 'distance_to_railway', 'school_count_1km',
       'kindergarten_count_1km', 'distance_to_supermarket',
       'distance_to_techhub', 'distance_to_park', 'distance_to_waterbody',
       'park_area_m2', 'waterbody_area_m2', 'Floor_clean', 'SaleType_encoded',
       'park_leisure_garden', 'park_leisure_park', 'park_leisure_pitch',
       'park_leisure_playground', 'park_leisure_recreation_common',
       'waterbody_type_lake', 'waterbody_type_pond',
       'waterbody_type_reservoir', 'waterbody_type_river', 'log_amount',
       'area_raw', 'log_area'],
      dtype='object')

df_dropped.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 4665 entries, 0 to 4664
Data columns (total 29 columns):
 #   Column                          Non-Null Count  Dtype  
---  ------                          --------------  -----  
 0   Latitude                        4665 non-null   float64
 1   Longitude                       4665 non-null   float64
 2   bus_stop_count_1km              4665 non-null   int64  
 3   financial_facilities_1km        4665 non-null   int64  
 4   medical_facilities_1km          4665 non-null   int64  
 5   distance_to_metro               4665 non-null   float64
 6   distance_to_railway             4665 non-null   float64
 7   school_count_1km                4665 non-null   int64  
 8   kindergarten_count_1km          4665 non-null   int64  
 9   distance_to_supermarket         4665 non-null   float64
 10  distance_to_techhub             4665 non-null   float64
 11  distance_to_park                4665 non-null   float64
 12  distance_to_waterbody           4665 non-null   float64
 13  park_area_m2                    4665 non-null   float64
 14  waterbody_area_m2               4665 non-null   float64
 15  Floor_clean                     4665 non-null   int64  
 16  SaleType_encoded                4665 non-null   int64  
 17  park_leisure_garden             4665 non-null   int32  
 18  park_leisure_park               4665 non-null   int32  
 19  park_leisure_pitch              4665 non-null   int32  
 20  park_leisure_playground         4665 non-null   int32  
 21  park_leisure_recreation_common  4665 non-null   int32  
 22  waterbody_type_lake             4665 non-null   int32  
 23  waterbody_type_pond             4665 non-null   int32  
 24  waterbody_type_reservoir        4665 non-null   int32  
 25  waterbody_type_river            4665 non-null   int32  
 26  log_amount                      4665 non-null   float64
 27  area_raw                        4665 non-null   float64
 28  log_area                        4665 non-null   float64
dtypes: float64(13), int32(9), int64(7)
memory usage: 893.0 KB

import pandas as pd
import geopandas as gpd
import numpy as np
from mgwr.gwr import GWR
from mgwr.sel_bw import Sel_BW
from mgwr.utils import shift_colormap

# --- Step 1: Convert to GeoDataFrame ---
gdf = gpd.GeoDataFrame(
    df_dropped,
    geometry=gpd.points_from_xy(df_dropped['Longitude'], df_dropped['Latitude']),
    crs='EPSG:4326'
)

# --- Step 2: Project to UTM (Bangalore ≈ EPSG:32643) ---
gdf_proj = gdf.to_crs(epsg=32643)

# --- Step 3: Define dependent and independent variables ---
y_var = 'log_amount'
X_vars = [
    'bus_stop_count_1km', 'financial_facilities_1km', 'medical_facilities_1km', 'distance_to_metro', 'distance_to_railway',
    'school_count_1km', 'kindergarten_count_1km', 'distance_to_supermarket',
    'distance_to_techhub', 'distance_to_park', 'distance_to_waterbody',
    'park_area_m2', 'waterbody_area_m2', 'Floor_clean', 'SaleType_encoded',
    'park_leisure_garden', 'park_leisure_park',
       'park_leisure_pitch', 'park_leisure_playground',
       'park_leisure_recreation_common',
       'waterbody_type_lake',
       'waterbody_type_pond', 'waterbody_type_reservoir',
       'waterbody_type_river', 'area_raw', 'log_area'
]

from statsmodels.stats.outliers_influence import variance_inflation_factor

X_df = gdf_proj[X_vars].copy()

# Compute VIF for each variable
vif_data = pd.DataFrame()
vif_data["feature"] = X_df.columns
vif_data["VIF"] = [variance_inflation_factor(X_df.values, i) for i in range(X_df.shape[1])]

vif_data.sort_values("VIF", ascending=False)

X_df.columns

Index(['bus_stop_count_1km', 'financial_facilities_1km',
       'medical_facilities_1km', 'distance_to_metro', 'distance_to_railway',
       'school_count_1km', 'kindergarten_count_1km', 'distance_to_supermarket',
       'distance_to_techhub', 'distance_to_park', 'distance_to_waterbody',
       'park_area_m2', 'waterbody_area_m2', 'Floor_clean', 'SaleType_encoded',
       'park_leisure_garden', 'park_leisure_park', 'park_leisure_pitch',
       'park_leisure_playground', 'park_leisure_recreation_common',
       'waterbody_type_lake', 'waterbody_type_pond',
       'waterbody_type_reservoir', 'waterbody_type_river', 'area_raw',
       'log_area'],
      dtype='object')

X_vars_gwr = [
       'financial_facilities_1km', 'medical_facilities_1km', 'distance_to_railway',
       'school_count_1km', 'kindergarten_count_1km',
       'distance_to_techhub', 'distance_to_park', 'distance_to_waterbody',
       'park_area_m2', 'Floor_clean', 'SaleType_encoded', 'log_area'
]

from mgwr.gwr import GWR
from mgwr.sel_bw import Sel_BW
import numpy as np

# --- Prepare Variables ---
y = np.array(gdf_proj[y_var]).reshape((-1, 1))
X = np.array(gdf_proj[X_vars_gwr])
coords = np.column_stack((gdf_proj.geometry.x, gdf_proj.geometry.y))

print(f"Data ready: {X.shape[0]} observations, {X.shape[1]} predictors")

# --- Step 1: Bandwidth Selection ---
bw = Sel_BW(coords, y, X, fixed=True, kernel='gaussian').search()
print(f"\n✅ Optimal Bandwidth: {bw:.2f}")

# --- Step 2: Fit GWR (with variable names included) ---
gwr_model = GWR(
    coords,
    y,
    X,
    bw,
    fixed=True,
    kernel='gaussian',
    constant=True,
    name_x=X_vars_gwr
).fit()

# --- Step 3: Print Model Summary ---
print("\n=========== GWR Summary with Variable Names ===========")
print(gwr_model.summary())

Data ready: 4665 observations, 12 predictors

✅ Optimal Bandwidth: 652.29

=========== GWR Summary with Variable Names ===========
===========================================================================
Model type                                                         Gaussian
Number of observations:                                                4665
Number of covariates:                                                    13

Global Regression Results
---------------------------------------------------------------------------
Residual sum of squares:                                           1010.221
Log-likelihood:                                                   -3050.813
AIC:                                                               6127.626
AICc:                                                              6129.716
BIC:                                                             -38289.145
R2:                                                                   0.243
Adj. R2:                                                              0.241

Variable                              Est.         SE  t(Est/SE)    p-value
------------------------------- ---------- ---------- ---------- ----------
Intercept                           12.511      0.130     96.431      0.000
financial_facilities_1km             0.010      0.001      9.047      0.000
medical_facilities_1km              -0.005      0.001     -4.312      0.000
distance_to_railway                 -0.000      0.000     -5.223      0.000
school_count_1km                    -0.010      0.002     -5.342      0.000
kindergarten_count_1km               0.011      0.003      3.435      0.001
distance_to_techhub                 -0.000      0.000     -4.760      0.000
distance_to_park                    -0.000      0.000    -10.815      0.000
distance_to_waterbody               -0.000      0.000     -2.837      0.005
park_area_m2                         0.000      0.000      2.290      0.022
Floor_clean                          0.000      0.000      2.208      0.027
SaleType_encoded                    -0.085      0.015     -5.487      0.000
log_area                             0.495      0.017     28.355      0.000

Geographically Weighted Regression (GWR) Results
---------------------------------------------------------------------------
Spatial kernel:                                          Fixed gaussian
Bandwidth used:                                                     652.290

Diagnostic information
---------------------------------------------------------------------------
Residual sum of squares:                                            319.433
Effective number of parameters (trace(S)):                         1165.218
Degree of freedom (n - trace(S)):                                  3499.782
Sigma estimate:                                                       0.302
Log-likelihood:                                                    -365.223
AIC:                                                               3062.882
AICc:                                                              3841.221
BIC:                                                              10582.470
R2:                                                                   0.761
Adjusted R2:                                                          0.681
Adj. alpha (95%):                                                     0.001
Adj. critical t value (95%):                                          3.454

Summary Statistics For GWR Parameter Estimates
---------------------------------------------------------------------------
Variable                   Mean        STD        Min     Median        Max
-------------------- ---------- ---------- ---------- ---------- ----------
Intercept                10.998      3.732    -27.470     11.120     21.980
financial_facilities      0.005      0.102     -1.104      0.002      1.673
medical_facilities_1      0.011      0.176     -1.461     -0.005      3.272
distance_to_railway      -0.000      0.001     -0.003     -0.000      0.005
school_count_1km         -0.004      0.113     -1.314     -0.001      0.500
kindergarten_count_1 -1304184.429 41627221.527 -1270660431.787     -0.003 78661905.723
distance_to_techhub      -0.000      0.001     -0.005     -0.000      0.004
distance_to_park         -0.000      0.001     -0.013     -0.000      0.007
distance_to_waterbod      0.000      0.001     -0.007     -0.000      0.004
park_area_m2              0.000      0.000     -0.000     -0.000      0.001
Floor_clean               0.013      0.023     -0.091      0.012      0.085
SaleType_encoded         -0.143      0.313     -2.598     -0.112      0.808
log_area                  0.704      0.466     -1.215      0.687      6.088
===========================================================================

None

import matplotlib.pyplot as plt
import geopandas as gpd

# --- Step 1: Add GWR coefficients to GeoDataFrame ---
# Add 'Intercept' manually since gwr_model itself doesn't store that flag
var_names = ['Intercept'] + X_vars_gwr if gwr_model.params.shape[1] == len(X_vars_gwr) + 1 else X_vars_gwr

for i, var in enumerate(var_names):
    gdf_proj[f'{var}_coef'] = gwr_model.params[:, i]

# --- Step 2: Select the variable to visualize ---
var_to_plot = 'distance_to_park_coef'  # adjust if your variable name differs

# --- Step 3: Plot heatmap of local coefficients ---
fig, ax = plt.subplots(figsize=(10, 8))

gdf_proj.plot(
    column=var_to_plot,
    cmap='coolwarm',
    legend=True,
    scheme='quantiles',
    k=5,
    alpha=0.9,
    ax=ax
)

ax.set_title("Local Effect of Distance to Green Space on Housing Prices (GWR)",
             fontsize=14, pad=12)
ax.axis('off')

plt.tight_layout()
plt.show()

# --- Step 1: Add GWR coefficients to GeoDataFrame ---
# Add 'Intercept' manually since gwr_model itself doesn't store that flag
var_names = ['Intercept'] + X_vars_gwr if gwr_model.params.shape[1] == len(X_vars_gwr) + 1 else X_vars_gwr

for i, var in enumerate(var_names):
    gdf_proj[f'{var}_coef'] = gwr_model.params[:, i]

# --- Step 2: Select the variable to visualize ---
var_to_plot = 'distance_to_waterbody_coef'  # adjust if your variable name differs

# --- Step 3: Plot heatmap of local coefficients ---
fig, ax = plt.subplots(figsize=(10, 8))

gdf_proj.plot(
    column=var_to_plot,
    cmap='coolwarm',
    legend=True,
    scheme='quantiles',
    k=5,
    alpha=0.9,
    ax=ax
)

ax.set_title("Local Effect of Distance to Blue Space on Housing Prices (GWR)",
             fontsize=14, pad=12)
ax.axis('off')

plt.tight_layout()
plt.show()

# Load BBMP wards or zones shapefile
bbmp = gpd.read_file("BBMP_shape.shp")

# Ensure same CRS as your transaction data
bbmp = bbmp.to_crs(gdf_proj.crs)

import contextily as ctx
import matplotlib.pyplot as plt
import geopandas as gpd

# --- Step 1: Add GWR coefficients to GeoDataFrame ---
# Add 'Intercept' manually since gwr_model itself doesn't store that flag
var_names = ['Intercept'] + X_vars_gwr if gwr_model.params.shape[1] == len(X_vars_gwr) + 1 else X_vars_gwr

for i, var in enumerate(var_names):
    gdf_proj[f'{var}_coef'] = gwr_model.params[:, i]

# --- Step 2: Select the variable to visualize ---
var_to_plot = 'distance_to_waterbody_coef'  # adjust if your variable name differs

# --- Step 2: Project both datasets to Web Mercator ---
gdf_proj_web = gdf_proj.to_crs(epsg=3857)
bbmp_web = bbmp.to_crs(epsg=3857)

# --- Step 3: Plot coefficient map (e.g., distance_to_park) ---
fig, ax = plt.subplots(figsize=(12, 10))

# Plot BBMP boundary
bbmp_web.boundary.plot(ax=ax, color='black', linewidth=0.5)

# Plot local coefficients (now it exists!)
gdf_proj_web.plot(
    column='distance_to_waterbody_coef',
    cmap='coolwarm',
    legend=True,
    scheme='quantiles',
    k=5,
    alpha=0.8,
    ax=ax
)

# Add OSM basemap
ctx.add_basemap(ax, source=ctx.providers.OpenStreetMap.Mapnik)

# Styling
ax.set_title("Local Effect of Distance to waterbody on Housing Prices (GWR)", fontsize=14)
ax.axis('off')

plt.show()

pip install matplotlib-scalebar

Requirement already satisfied: matplotlib-scalebar in c:\users\sayan\anaconda3\lib\site-packages (0.9.0)
Requirement already satisfied: matplotlib in c:\users\sayan\anaconda3\lib\site-packages (from matplotlib-scalebar) (3.9.2)
Requirement already satisfied: contourpy>=1.0.1 in c:\users\sayan\anaconda3\lib\site-packages (from matplotlib->matplotlib-scalebar) (1.2.0)
Requirement already satisfied: cycler>=0.10 in c:\users\sayan\anaconda3\lib\site-packages (from matplotlib->matplotlib-scalebar) (0.11.0)
Requirement already satisfied: fonttools>=4.22.0 in c:\users\sayan\anaconda3\lib\site-packages (from matplotlib->matplotlib-scalebar) (4.51.0)
Requirement already satisfied: kiwisolver>=1.3.1 in c:\users\sayan\anaconda3\lib\site-packages (from matplotlib->matplotlib-scalebar) (1.4.4)
Requirement already satisfied: numpy>=1.23 in c:\users\sayan\anaconda3\lib\site-packages (from matplotlib->matplotlib-scalebar) (1.26.4)
Requirement already satisfied: packaging>=20.0 in c:\users\sayan\anaconda3\lib\site-packages (from matplotlib->matplotlib-scalebar) (24.1)
Requirement already satisfied: pillow>=8 in c:\users\sayan\anaconda3\lib\site-packages (from matplotlib->matplotlib-scalebar) (10.4.0)
Requirement already satisfied: pyparsing>=2.3.1 in c:\users\sayan\anaconda3\lib\site-packages (from matplotlib->matplotlib-scalebar) (3.1.2)
Requirement already satisfied: python-dateutil>=2.7 in c:\users\sayan\anaconda3\lib\site-packages (from matplotlib->matplotlib-scalebar) (2.9.0.post0)
Requirement already satisfied: six>=1.5 in c:\users\sayan\anaconda3\lib\site-packages (from python-dateutil>=2.7->matplotlib->matplotlib-scalebar) (1.16.0)
Note: you may need to restart the kernel to use updated packages.

import contextily as ctx
import matplotlib.pyplot as plt
from matplotlib_scalebar.scalebar import ScaleBar

# Project both datasets to Web Mercator for contextily
gdf_proj_web = gdf_proj.to_crs(epsg=3857)
bbmp_web = bbmp.to_crs(epsg=3857)

# Plot coefficient map (e.g., distance_to_park)
fig, ax = plt.subplots(figsize=(12, 10))

# Plot BBMP boundary
bbmp_web.boundary.plot(ax=ax, color='black', linewidth=0.5)

# Plot local coefficients
gdf_proj_web.plot(column='distance_to_waterbody_coef',
                  cmap='coolwarm', legend=True,
                  scheme='quantiles', k=5,  # optional classification
                  alpha=0.8, ax=ax)

# Add OSM basemap
ctx.add_basemap(ax, source=ctx.providers.CartoDB.Positron)

# --- Add map elements ---
ax.set_title("Local Effect of Distance to Waterbody on Housing Prices (GWR)", fontsize=16)
ax.axis('off')

# Add north arrow
ax.text(0.95, 0.1, 'N', transform=ax.transAxes, ha='center', va='center',
        fontsize=16, fontweight='bold')
ax.arrow(0.95, 0.12, 0, 0.05, transform=ax.transAxes, color='k',
         head_width=0.02, head_length=0.03)

# Add scale bar (using meters as CRS units)
scalebar = ScaleBar(1, units="m", location='lower left')
ax.add_artist(scalebar)

# --- Save high-resolution image ---
output_path = "outputs/gwr_distance_to_waterbody.png"
plt.savefig(output_path, dpi=600, bbox_inches='tight')
plt.close()

print(f"✅ Map saved successfully to: {output_path}")

✅ Map saved successfully to: outputs/gwr_distance_to_waterbody.png

pip install matplotlib-scalebar

Requirement already satisfied: matplotlib-scalebar in c:\users\sayan\anaconda3\lib\site-packages (0.9.0)
Requirement already satisfied: matplotlib in c:\users\sayan\anaconda3\lib\site-packages (from matplotlib-scalebar) (3.9.2)
Requirement already satisfied: contourpy>=1.0.1 in c:\users\sayan\anaconda3\lib\site-packages (from matplotlib->matplotlib-scalebar) (1.2.0)
Requirement already satisfied: cycler>=0.10 in c:\users\sayan\anaconda3\lib\site-packages (from matplotlib->matplotlib-scalebar) (0.11.0)
Requirement already satisfied: fonttools>=4.22.0 in c:\users\sayan\anaconda3\lib\site-packages (from matplotlib->matplotlib-scalebar) (4.51.0)
Requirement already satisfied: kiwisolver>=1.3.1 in c:\users\sayan\anaconda3\lib\site-packages (from matplotlib->matplotlib-scalebar) (1.4.4)
Requirement already satisfied: numpy>=1.23 in c:\users\sayan\anaconda3\lib\site-packages (from matplotlib->matplotlib-scalebar) (1.26.4)
Requirement already satisfied: packaging>=20.0 in c:\users\sayan\anaconda3\lib\site-packages (from matplotlib->matplotlib-scalebar) (24.1)
Requirement already satisfied: pillow>=8 in c:\users\sayan\anaconda3\lib\site-packages (from matplotlib->matplotlib-scalebar) (10.4.0)
Requirement already satisfied: pyparsing>=2.3.1 in c:\users\sayan\anaconda3\lib\site-packages (from matplotlib->matplotlib-scalebar) (3.1.2)
Requirement already satisfied: python-dateutil>=2.7 in c:\users\sayan\anaconda3\lib\site-packages (from matplotlib->matplotlib-scalebar) (2.9.0.post0)
Requirement already satisfied: six>=1.5 in c:\users\sayan\anaconda3\lib\site-packages (from python-dateutil>=2.7->matplotlib->matplotlib-scalebar) (1.16.0)
Note: you may need to restart the kernel to use updated packages.

'''
import os
import contextily as ctx
import matplotlib.pyplot as plt
from matplotlib_scalebar.scalebar import ScaleBar

# --- Create output folder if it doesn’t exist ---
os.makedirs("outputs", exist_ok=True)

# --- Reproject for web mapping tiles ---
gdf_proj_web = gdf_proj.to_crs(epsg=3857)
bbmp_web = bbmp.to_crs(epsg=3857)

# --- Plot setup ---
fig, ax = plt.subplots(figsize=(12, 10))

# Plot BBMP boundary
bbmp_web.boundary.plot(ax=ax, color='black', linewidth=0.5)

# Plot GWR coefficient map (example variable)
gdf_proj_web.plot(
    column='distance_to_park_coef',
    cmap='coolwarm',
    legend=True,
    scheme='quantiles', k=5,
    alpha=0.8,
    ax=ax
)

# Add OSM basemap
ctx.add_basemap(ax, source=ctx.providers.CartoDB.Positron)

# --- Add map elements ---
ax.set_title("Local Effect of Distance to Park on Housing Prices (GWR)", fontsize=16)
ax.axis('off')

# Add north arrow
ax.text(0.95, 0.1, 'N', transform=ax.transAxes, ha='center', va='center',
        fontsize=16, fontweight='bold')
ax.arrow(0.95, 0.12, 0, 0.05, transform=ax.transAxes, color='k',
         head_width=0.02, head_length=0.03)

# Add scale bar (using meters as CRS units)
scalebar = ScaleBar(1, units="m", location='lower left')
ax.add_artist(scalebar)

# --- Save high-resolution image ---
output_path = "outputs/gwr_distance_to_park.png"
plt.savefig(output_path, dpi=300, bbox_inches='tight')
plt.close()

print(f"✅ Map saved successfully to: {output_path}")
'''

'\nimport os\nimport contextily as ctx\nimport matplotlib.pyplot as plt\nfrom matplotlib_scalebar.scalebar import ScaleBar\n\n# --- Create output folder if it doesn’t exist ---\nos.makedirs("outputs", exist_ok=True)\n\n# --- Reproject for web mapping tiles ---\ngdf_proj_web = gdf_proj.to_crs(epsg=3857)\nbbmp_web = bbmp.to_crs(epsg=3857)\n\n# --- Plot setup ---\nfig, ax = plt.subplots(figsize=(12, 10))\n\n# Plot BBMP boundary\nbbmp_web.boundary.plot(ax=ax, color=\'black\', linewidth=0.5)\n\n# Plot GWR coefficient map (example variable)\ngdf_proj_web.plot(\n    column=\'distance_to_park_coef\',\n    cmap=\'coolwarm\',\n    legend=True,\n    scheme=\'quantiles\', k=5,\n    alpha=0.8,\n    ax=ax\n)\n\n# Add OSM basemap\nctx.add_basemap(ax, source=ctx.providers.CartoDB.Positron)\n\n# --- Add map elements ---\nax.set_title("Local Effect of Distance to Park on Housing Prices (GWR)", fontsize=16)\nax.axis(\'off\')\n\n# Add north arrow\nax.text(0.95, 0.1, \'N\', transform=ax.transAxes, ha=\'center\', va=\'center\',\n        fontsize=16, fontweight=\'bold\')\nax.arrow(0.95, 0.12, 0, 0.05, transform=ax.transAxes, color=\'k\',\n         head_width=0.02, head_length=0.03)\n\n# Add scale bar (using meters as CRS units)\nscalebar = ScaleBar(1, units="m", location=\'lower left\')\nax.add_artist(scalebar)\n\n# --- Save high-resolution image ---\noutput_path = "outputs/gwr_distance_to_park.png"\nplt.savefig(output_path, dpi=300, bbox_inches=\'tight\')\nplt.close()\n\nprint(f"✅ Map saved successfully to: {output_path}")\n'

# --- Step 1: Add local R² values to your GeoDataFrame ---
# gwr_model.localR2 gives an array of local R² for each observation
gdf_proj['local_R2'] = gwr_model.localR2

# --- Step 2: Project both datasets to Web Mercator ---
gdf_proj_web = gdf_proj.to_crs(epsg=3857)
bbmp_web = bbmp.to_crs(epsg=3857)

# --- Step 3: Plot Local R² map ---
fig, ax = plt.subplots(figsize=(12, 10))

# Plot BBMP boundary
bbmp_web.boundary.plot(ax=ax, color='grey', linewidth=0.5)

# Plot local R²
gdf_proj_web.plot(
    column='local_R2',
    cmap='viridis',
    legend=True,
    alpha=0.8,
    scheme='quantiles',  # optional for clearer breaks
    k=5,
    ax=ax
)

# Add light basemap
ctx.add_basemap(ax, source=ctx.providers.CartoDB.Positron)

# Style
ax.set_title("Local R² from GWR Model", fontsize=14, pad=12)
ax.axis('off')

plt.tight_layout()
plt.show()

'''
import os
import contextily as ctx
import matplotlib.pyplot as plt

# --- Create output folder if it doesn’t exist ---
os.makedirs("outputs", exist_ok=True)

# --- Plot Local R² ---
fig, ax = plt.subplots(figsize=(12, 10))

# Plot BBMP boundary
bbmp_web.boundary.plot(ax=ax, color='grey', linewidth=0.5)

# Plot local R² values
gdf_proj_web.plot(
    column='local_R2',
    cmap='viridis',
    legend=True,
    alpha=0.8,
    ax=ax
)

# Add OSM basemap
ctx.add_basemap(ax, source=ctx.providers.CartoDB.Positron)

# Style and title
ax.set_title("Local R² from GWR Model", fontsize=14)
ax.axis('off')

# --- Save image ---
output_path = "outputs/gwr_local_R2_map.png"
plt.savefig(output_path, dpi=300, bbox_inches='tight')
plt.close()

print(f"✅ Map saved successfully at: {output_path}")
'''

'\nimport os\nimport contextily as ctx\nimport matplotlib.pyplot as plt\n\n# --- Create output folder if it doesn’t exist ---\nos.makedirs("outputs", exist_ok=True)\n\n# --- Plot Local R² ---\nfig, ax = plt.subplots(figsize=(12, 10))\n\n# Plot BBMP boundary\nbbmp_web.boundary.plot(ax=ax, color=\'grey\', linewidth=0.5)\n\n# Plot local R² values\ngdf_proj_web.plot(\n    column=\'local_R2\',\n    cmap=\'viridis\',\n    legend=True,\n    alpha=0.8,\n    ax=ax\n)\n\n# Add OSM basemap\nctx.add_basemap(ax, source=ctx.providers.CartoDB.Positron)\n\n# Style and title\nax.set_title("Local R² from GWR Model", fontsize=14)\nax.axis(\'off\')\n\n# --- Save image ---\noutput_path = "outputs/gwr_local_R2_map.png"\nplt.savefig(output_path, dpi=300, bbox_inches=\'tight\')\nplt.close()\n\nprint(f"✅ Map saved successfully at: {output_path}")\n'

from platform import python_version
print(python_version())

3.12.7

	log_area	Floor_clean	SaleType_encoded	bus_stop_count_1km	financial_facilities_1km	medical_facilities_1km	distance_to_metro	distance_to_railway	school_count_1km	kindergarten_count_1km	distance_to_supermarket	distance_to_techhub	distance_to_park	distance_to_waterbody	park_area_m2	waterbody_area_m2
0	7.117206	3	0	3	0	1	1021.855251	1031.314063	0	0	1071.728427	5012.910521	205.823897	963.347727	11609.389521	15481.502122
1	7.632401	4	0	7	7	8	1624.793018	1640.439676	2	1	324.707763	1371.896658	215.581180	816.380253	1001.151224	403075.658943
2	7.901007	4	1	7	7	5	2323.930856	2343.893908	7	0	621.209341	2633.436246	560.260318	1138.757111	518.173805	12577.559117
3	7.306531	4	0	8	12	6	71.024054	87.654521	6	1	144.297691	562.039276	128.222356	671.607186	147.420732	25026.590223
4	7.841886	20	0	2	0	0	1789.025763	1805.479717	4	0	1653.838238	3822.477586	70.367519	1359.177291	640.230205	163020.728866
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
4660	7.801391	16	0	12	6	1	7410.102912	2553.008614	4	0	1256.031510	879.973099	238.373796	508.605424	859.862494	383.302762
4661	7.829233	26	0	12	6	1	7410.102912	2553.008614	4	0	1256.031510	879.973099	238.373796	508.605424	859.862494	383.302762
4662	6.806829	9	0	9	11	6	832.198921	853.351181	3	0	386.751500	768.605871	157.895705	1122.355482	503.926172	28561.029058
4663	7.254178	0	0	9	11	6	832.198921	853.351181	3	0	386.751500	768.605871	157.895705	1122.355482	503.926172	28561.029058
4664	7.783224	2	0	9	11	6	832.198921	853.351181	3	0	386.751500	768.605871	157.895705	1122.355482	503.926172	28561.029058

Data Preparation¶

OLS model¶

LISA MAP¶

SEM model results¶

SLM model¶

GWR¶

	feature	VIF
25	log_area	349.771278
20	waterbody_type_lake	144.649901
21	waterbody_type_pond	51.974327
22	waterbody_type_reservoir	28.310408
24	area_raw	25.577184
23	waterbody_type_river	8.556582
1	financial_facilities_1km	7.189323
4	distance_to_railway	5.851528
2	medical_facilities_1km	5.672754
0	bus_stop_count_1km	5.400138
8	distance_to_techhub	5.212559
5	school_count_1km	4.881945
10	distance_to_waterbody	4.832278
7	distance_to_supermarket	4.521687
16	park_leisure_park	4.266484
3	distance_to_metro	3.805006
17	park_leisure_pitch	3.789340
9	distance_to_park	3.725549
6	kindergarten_count_1km	1.933640
18	park_leisure_playground	1.826687
15	park_leisure_garden	1.603634
12	waterbody_area_m2	1.492221
14	SaleType_encoded	1.415572
11	park_area_m2	1.244951
19	park_leisure_recreation_common	1.086052
13	Floor_clean	1.006822